BACKGROUND IMAGE: iSTOCK/GETTY IMAGES
Editor's note: Jim Fowler became CIO at Nationwide Mutual Insurance Co. in July 2018.
As CIO at GE, Jim Fowler pushed his colleagues to see him as something more than the guy who keeps the lights on and the help desk tickets flowing. Instead, he asked the business leaders to consider him the chief automation officer.
In this video interview filmed at the 2017 MIT Sloan CIO Symposium in Cambridge, Mass., Fowler described what he sees as the primary responsibilities of a chief automation officer. He also talked about how artificial intelligence (AI), machine learning and the company's Predix platform, which collects and analyzes data from the industrial internet of things, helped him fulfill the role.
Who owns the AI budget at GE?
Jim Fowler: Inside GE -- for the GE component of AI -- that's in my budget. There's also a GE for customer component where we've made an acquisition this year in the AI space. And that actually sits in the commercial team's budget. I'm their biggest customer. I own the spending that we have there, as well as with other AI vendors.
What are you spending that budget on?
Fowler: We're looking at how we can automate. That's the other big change happening inside [IT's] role: I've told my business leaders to stop thinking about the CIO as the person running the network, the PCs, etc. Those are table stakes. Instead, think about this person as the chief automation officer who should be helping point out how work is going to change over the next 10 years, how machines are going to tell people what to do more than people telling machines what to do, and how they are going to help you see where productivity will come from as we go through that automation market. AI and machine learning are at the heart of all of that.
How are you measuring the ROI on those investments?
Fowler: For me, it's broader than artificial intelligence -- it's technology in general. And when I look at the application of technology inside the company, the measures are pretty simple. There are three metrics that my variable compensation is based on: The first is operating profit driven by productivity; I have a $700 million target for productivity this year that I have to meet. There's also a free cash flow metric that I have to meet, and there's a revenue generation and orders metric that I have to meet.
It's those business metrics that determine the success of a program. If I can't tie a program back to one of those three things, we, frankly, shouldn't be working on it.
What steps are you taking to hit that $700 million productivity target?
Fowler: I'll give you an example: Our services business is probably one of the biggest elements of productivity for us. If you picture GE, we sell big equipment -- aircraft engines, locomotives, turbines, MRI units. We have a large field staff -- 40,000 field workers that are out doing field maintenance and support for our customers. We've implemented a set of technologies over this last year that starts to automate and simplify their world.
In our power business, field service engineers use 86 different applications to do their job. Think about the process complexity behind all that. We've used our Predix platform to build one application that's persona-based that's focused on how they do their job; that gives them every transaction, every piece of data, every drawing that they need to do their job. We've started to overlay augmented reality on top of visualizations to show them how to do their job. And we've used collaboration tools to let them collaborate and see how the assets that they're working on are running in real time. That alone [resulted in] $200 million of productivity last year.
We look at each job as a persona, and we look at how can to make that persona -- that person's job -- easier to do.